Image processing apparatus and image processing method

Abstract

An image scanner using an area sensor having a tilt reads a plurality of low-resolution image data having phase shifts from each other, and the low-resolution image data are converted into those on an orthogonal coordinate system by affine transformation. The number of data to be used is decided based on one of these low-resolution image data. The low-resolution image data as many as the designated number of data are saved, and high-resolution image data is generated by executing super-resolution processing.

Description

BACKGROUND OF THE INVENTION[0001]1. Field of the Invention[0002]The present invention relates to an image processing apparatus, image processing method and a program, which execute copy processing and transmission processing using, for example, a scanner and, more particularly, to an image processing apparatus, image processing method and a program, which use a super-resolution technique.[0003]2. Description of the Related Art[0004]A technique called “super-resolution”, which enhances a resolution using a plurality of images having a given resolution, is known. With this technique, an image of 600 dpi can be converted into an image of 1200 dpi or the equivalent, and a conventional device can be used to obtain an image having a higher resolution. In the present specification, when an image is converted into pixels, the number of pixels (i.e., a pixel density) per unit length or per unit area of that image is called a resolution of the image. In order to attain the super-resolution te...

AI Technical Summary

Benefits of technology

The present invention provides an image processing apparatus and method that can determine the necessity of high-resolution processing with high precision to avoid execution of super-resolution processing with low necessity and to adequately execute super-resolution processing with high necessity. The apparatus includes a decision unit and a super-resolution processing unit that can decide whether or not to execute super-resolution processing based on frequency characteristics of the low-resolution image data. The super-resolution processing unit can generate a high-resolution image by synthesizing a plurality of low-resolution image data obtained from an identical image and have phase shifts from each other. The apparatus can prevent performance drop and increase in required memory size caused by super-resolution processing and implement image processing that can assure high image quality.

Problems solved by technology

However, since the super-resolution processing requires a plurality of images per pixel, the required memory size and calculation volume increase.

However, whether or not high-resolution processing is required for an input image of a multi-functional peripheral equipment (MFP) cannot be determined based only on the size of an area of interest and luminance difference.

For example, when a complicated image such as a map or an image of a brochure including many fine characters are input, they undergo low-resolution processing in the related art, and a problem of crushed small characters is posed even though it has been decided that luminance values are not different in a plurality of images.

When OCR processing is executed for an input image that has a low resolution, particularly, characters of small points are recognized as wrong characters or not recognized, resulting in low character recognition precision.

Images